Characterizing Reading Comprehension Ability: Discrete Strategies vs. A Holistic Approach

Article by Paolo Martin

The act of comprehending text is complex.  The various strategies for comprehending text are often difficult to characterize, as every individual who interacts with a text brings a different set of prior experiences, reasons for reading the text, and strategies for understanding it.

Comprehension assessments exist in many different forms.  Some assessments simply try to determine a student's reading level.  Other assessments attempt to break down comprehension into discrete comprehension strategies such as identifying main ideas, using background knowledge, monitoring comprehension, predicting, generating questions, summarizing, etc.  Assessments that attempt to break down the reading comprehension process for kids, in theory, help teachers identify the "holes" in children's comprehension strategies and allow teachers to align their reading instruction to state or district comprehension standards, which are sometimes broken down into discrete strategies. 

While there are advantages to attempting to identify discrete deficiencies in a child's comprehension ability, there also exist real issues with this process.  First, the definitions of the many widely accepted comprehension strategies are constantly debated by educators and academics and therefore are difficult to pin down.  For example, while "monitoring comprehension" is a widely accepted and valued comprehension strategy, it is often debated what distinguishes "monitoring comprehension" from "using background knowledge," or "generating questions," or "identifying the author's purpose."  Some argue that if children use "background knowledge" or "generate questions," they are indeed "monitoring comprehension" and perhaps formulating the "main idea."  Similarly, some argue that certain comprehension strategies are really a subset of other comprehension strategies. 

A second difficulty with attempting to break down comprehension into discrete strategies in an assessment setting is being able to accurately come to a conclusion about which strategies really are problematic for the child.  This is related to difficulties not only in defining the discrete comprehension strategies that are often proposed, but also in asking enough of a particular kind of comprehension strategy question from the multitude of comprehension strategies to accurately say that a specific problem exists.  While certain assessment questions in some standardized tests, for example, may be labeled as "main idea" or "using background knowledge" questions, it is still difficult to accurately conclude that students are or are not effectively using these skills.  The complexity of how children use strategies is acknowledged by the National Reading Panel in their description of good readers as being purposeful and active:

Good readers are purposeful.  Good readers have a purpose for reading. They may read to find out how to use a food processor, read a guidebook to gather information about national parks, read a textbook to satisfy the requirements of a course, read a magazine for entertainment, or read a classic novel to experience the pleasures of great literature.

Good readers are active.  Good readers think actively as they read. To make sense of what they read, good readers engage in a complicated process. Using their experience and knowledge of the world, their knowledge of vocabulary and language structure, and their knowledge of reading strategies (or plans), good readers make sense of the text and know how to get the most out of it. They know when they have problems with understanding and how to resolve these problems as they occur.

As such, assessments which attempt to measure a child's ability to comprehend using discrete tasks often do not take into account the child's individual, purposeful, and active  ways of understanding a piece of text.  For example, one child may find it necessary to ask a predictive question at one point in a passage, but another might not find it useful to do so to understand the same point in the same passage. Yet those tests might find one child less capable of comprehending than another child who simply uses his or her own comprehension strategy to understand the text.  According to the National Reading Panel, "...good readers must be able to coordinate and adjust several strategies to assist comprehension...."  The ways children adjust and manipulate these strategies are often very personal and complex and normally not validated in comprehension assessments which attempt to identify a deficiency in one strategy over another.

Let's Go Learn's Silent Reading sub-test, which is part of the Diagnostic Online Reading Assessment (DORA), does not attempt to find a student's deficiencies in any one particular comprehension strategy, for the reasons mentioned above.  Instead, it attempts to analyze the holistic ability of a child to understand a passage of a particular readability level and a grade-appropriate topic.  The questions' answers require the student either to recall an important detail from the passage or to make an inference about key concepts.  Thus, questions are put into only two general categories of "factual" or "inferential." The online assessment keeps administering passages of varying difficulty depending on the success of the child.  Additionally, the comprehension sub-test is not interpreted in isolation from the other sub-tests.  Instead, the DORA report generated for a particular child takes into account his or her decoding and vocabulary skills when commenting on the factors that may influence his or her ability to successfully comprehend a passage of a certain difficulty level. 

DORA is unique compared to many other assessments insofar it combines multiple measures to create an individualized prescriptive student profile.  DORA acknowledges that an accurate overview of a child's reading profile can only be generated by considering different factors that affect reading ability as opposed to characterizing reading ability by looking at subskills in isolation.  Additionally, DORA does not over-analyze comprehension strategies in isolation to come up with conclusions that are based on insignificant or confounded data.

> Previous page